The difference between voice playback chip and voice recognition chip

Voice chips are used in many products, such as voice playback chips, Bluetooth recording chips, music chips, etc. Although most people do not know the underlying principles of the chip, they basically have an understanding of the function of the chip. In order to enhance everyone’s understanding of chips, this article will introduce the voice playback chip and voice control chip.

The historical development of voice chips has evolved continuously by following the development needs of the times and the market. Geyuan Electronics is a leading company in the voice chip industry and has made indelible contributions in the development of the current programmable voice chips. Contribution, the company integrates R&D and production, constantly summarizes user needs from the market and continuously improves products, from voice chips, recording chips, MP3 music chips, voice modules, recording modules to the current voice recognition chip (voice control chip).

Voice playback chip

The voice playback chip has developed from the original mask voice chip to the current OTP voice chip that can be programmed with a programmer and the reprogrammable voice chip.

The sound produced by the voice chip is determined by the sound programmed into the chip. Then when it is applied to the circuit board, there are some buttons or the program writes the playback mode to trigger the playback of the sound.

That is to write and burn the underlying program for reading/writing into the programmable voice chip. Of course, you can also burn the sound files into it together. If the program supports it, you can also add an external memory to achieve the function of sound storage expansion.

There are various types of voice playback chips, and the methods used are also different. For example, OTP voice chips are disposable and reprogrammable voice chips are programmable. The one-time voice chip cannot be changed after being written, and the program and sound file are passed through once. The erasable voice chip can be reprogrammed multiple times, and the program can be modified and tested later.

Voice recognition chip (voice control chip)

Voice recognition chip is also called voice control IC. It is just one type of voice chip. It is mostly used for human-computer interaction. Whether it is smart home appliances or toys, they are developing in a more humane direction, so the application of intelligent voice recognition chip Very marketable. Compared with traditional voice chips, the biggest feature of voice recognition chips is its ability to recognize voices, which allows machines to understand human voices and perform various actions according to commands.

According to user restrictions, voice recognition chips can be divided into specific person voice recognition chips and non-specific person voice recognition chips.

Specific person speech recognition

The specific person speech recognition chip is for speech recognition of a designated person. It does not recognize the words of other people. The user’s speech reference sample must first be stored in the database for comparison. That is, the specific person speech recognition must undergo speech training before use. , generally follow the machine prompts to train the voice entries 2 times before use.

Non-specific speech recognition

Non-specific speech recognition is a recognition technology that does not target specific people, regardless of age or gender, as long as they speak the same language. The application model is to collect about 200 people based on a dozen or so voice interaction entries before the product is finalized. The sound samples are processed by PC algorithms to obtain the speech model and feature database of the interactive terms, and then burned into the chip. Machines (smart dolls, electronic pets, children’s computers) using this chip will have interactive functions.

Some non-person-specific speech recognition applications are phoneme-based algorithms. In this mode, interactive recognition can be performed without collecting many people’s voice samples. However, the disadvantage is that the recognition rate is not high and the recognition performance is unstable.

According to the continuity of speaking methods, speech recognition chips can be divided into discontinuous speech recognition and continuous speech recognition.

The most widely used in daily life is the non-specific speech recognition chip. The non-specific speech recognition chip is divided into offline speech chip and online speech chip according to whether it is connected to the Internet.

There is a difference between online and offline voice control chips:

• Offline speech recognition:fixed entries, no need to connect to the network, but the recognition rate is slightly lower
• Online speech recognition:The terms are not fixed and need to be connected to the network. The recognition rate is high, but the effect will be affected by the network and the price is relatively high.

Each technology has its application value, but different technologies are suitable for different fields. The offline recognition effect is slightly worse, but in a short distance and relatively quiet environment, the recognition rate can reach more than 90%; for some products that are not connected to the network , such as mobile lighting, massagers, etc. Offline speech recognition is more suitable than online and can meet many occasions; moreover, in terms of price, offline speech recognition is lower than online, and the corresponding products used in online speech recognition are also cheaper than offline. High, the rest depends on the market positioning and choice of customer service.